Central Pain as a Thalamocortical Dysrhythmia: A Thalamic Efference Disconnection?

The intrinsic electrical properties of neurons are presently considered as a salient parameter in brain function (Llinas 1988; Getting 1989; Connors and Gutnick 1990; Turrigiano et al. 1994; Margolis and Detwiler 2007). This is in contrast to the classical purely reflexological view, where neurons are considered to be passive agents that are activated or inhibited synaptically. This intrinsic functional view has been addressed in recent years in relation to thalamic neuron function and to its recurrent interaction with the cortex. Such a view is based upon single-cell neuronal electrophysiology (c.f. Llinas 1988; Steriade and Llinas 1988), and thalamocortical anatomy (Jones 2007). The neurological consequences of such a perspective (Llinas et al. 1999) have been corroborated by magnetoencephalography (MEG), single-cell intraoperative recordings, and encouraging surgical outcomes (Jeanmonod et al. 1996, 2001b, 2003). Moreover, as in the case of tinnitus, where a sound stimulus can suppress the centrally generated sensation (Coles and Hallam 1987), central pain can be modulated by peripheral stimulus (Somers and Somers 1999; Inui et al. 2006). From a functional imaging perspective, electroencephalogram (EEG) (Jeanmonod et al. 1993) and MEG data have shown the presence of a distinct increase in low-frequency activation in central pain (Schulman et al. 2005). In all patients with central, but not peripheral, pain, there was a second site of low-frequency oscillations that was localized to the mesial/orbito frontal and anterior cingulated cortices, as well as the temporal (insular) cortex. Central pain patients with these MEG characteristics did not respond to spinal cord stimulation. By contrast, patients without the frontal low-frequency component responded well to such stimulation. In addition to clinical studies, direct experimental evidence for the functional organization of the thalamo-cortico-thalamo loop has been obtained in studies of rodent thalamocortical slices (Llinas et al. 2002) that have been extended to animal studies concerning neuropathic pain (Gerke et al. 2003; Kim et al. 2003) and thalamic deafferentation (Wang and Thompson 2008). These latter results have established a direct relationship between abnormal thalamic rhythmicity and the occurrence of central pain. The relevance of an “essential thalamic structure” (i.e., a central generator) for neuropathic pain was first suggested by Head (Head and Holmes 1911). The findings summarized here extend this original proposal by addressing brain activity obtained from MEG, EEG, and preoperative unit recordings from patients with chronic neuropathic pain. We also briefly touch upon the contribution of animal studies to understanding the cellular and molecular components of neuropathic pain generation in the context of increased T-type calcium channel activty.